Rewetting, or quenching, is the re-establishment of liquid contact with a solid surface whose initial temperature exceeds the rewetting temperature (RT) — the maximal value for which the surface may wet.
Rewetting phenomena and rewetting models are reviewed by, e.g., Butterworth and Owen (1975), Elias and Yadigaroglu (1978), Carbajo and Siegel (1980), and Olek (1988).
Heat removal from a hot surface by a liquid may be hindered if the surface temperature (ST) is high enough to allow the formation of a vapor layer, acting as an insulator. When the ST drops below the RT, heat transfer can be enhanced by as much as two orders of magnitude.
The interest in rewetting of hot surfaces is mainly due to its role in light water reactor safety (see Blowdown and Reflood.) Following a hypothetical loss-of-coolant accident, the reactor core might become uncovered, and rapid cooling is needed. This is done either by top spraying of coolant or by bottom flooding. In either case, the task of the emergency core cooling system is to rewet the hot cladding of the fuel rods. Rewetting is also applicable to several industrial processes, such as metallurgical quenching of hot solids, start-up of LNG pipelines and filling of containers with cryogenic liquids. The terms quenching, sputtering, minimum film boiling point, return to nucleate boiling, departure from film boiling, film boiling collapse, and Leidenfrost Phenomenon have been used interchangeably to refer to various forms of rewetting. These terms, however, are not exactly synonymous.
Rewetting may occur in very different situations; a classification of the most frequent ones was made by Groeneveld (1984), Figure 1. The first four types refer to forced convective rewetting while the last two types deal with no-flow conditions. The different rewetting situations are described below. (See also Boiling.)
In inverted annular film boiling, there is a wavy liquid-vapor interface near the wall. If the wall temperature or heat flux are lowered, the vapor thickness decreases and eventually the liquid may contact the wall. Depending on the temperature level and the rate of heat supply to the cooled region beneath the liquid, permanent liquid-wall contact is either maintained or the liquid is pushed away from the surface with the formation of vapor. In the latter case, each contact is equivalent to an impulsive cooling of the surface. In a quenching process, repeated contacts will lower the ST enough to permit rewet.
Liquid contact with the heated surface is established by means of an axially propagating liquid film. At the quench front there is a narrow zone of vigorous boiling with production of an aerosol of droplets with diameters 0.5 mm and less. Some of the finer droplets are ejected almost horizontally from the film, which lifts off the hot surface at the quench front, causing the liquid to break up into large irregular droplets. At low liquid flows the liquid runs down the surface as a rivulet rather than as a continuous film. Here typical rewetting velocities are of the order of millimeters per second.
Axial conduction is usually the main mechanism of reducing the ST just ahead of the quench front, although precursory cooling by vapor and droplets ahead of the front may become important at higher flow rates.
Rewetting is due to an axially propagating quench front, with a region of nucleate and transition boiling behind it (Barnea et al. 1994). Different descriptions are reported regarding the behavior downstream of the quench front. In some cases the excess fluid forms a dispersed flow region of liquid and vapor coupled with film boiling. In other cases there exists an extensive film boiling region ahead of the quench front which breaks down to a mist flow at a distance of the order 10 to 200 mm. Rewetting velocities are of the order of centimeters per second. Prior to rewetting, the heated surface is precooled by transition and film boiling (inverted annular film boiling or dispersed flow film boiling) and/or axial conduction. At very low flow rates and high liquid subcoolings the precooling effect is expected to be insignificant.
The heated surface is precooled by a spray of droplets which can eventually lead to droplet rewetting. The rewetting depends on droplet size and velocity, vapor conditions and heated surface properties. Rewetting here is characterized by the absence of a quench front.
For type 5, the liquid is dispersed in the continuous gas phase, while for type 6, the gas forms a film on the surface; the liquid phase is continuous.
An additional form of rewetting, namely, rewetting by large masses of liquid without progression of a quench front, was reported by Aksan (1988).
To reflect the diversity of the rewetting phenomena, an indicative glossary is introduced for terms most often used to describe the temperature variation of a hot surface during its quenching. Note that this glossary is somewhat arbitrary and that these terms may be encountered in the literature under other meanings than those described here [Gerweck (1989)].
Rewetting temperature probably the most general term which, interpreted directly, refers to re-establishment of direct contact between wall and liquid. However, there is experimental evidence of short time contacts even in stable film boiling. There is another specific meaning for the rewetting temperature used in heat conduction models: that at which the heat transfer coefficient (assumed to vary axially) drops from nucleate boiling values to film boiling ones, for a rod being quenched in a reflood process.
Quench temperature is the value of the ST just before the temperature drops by the quenching process.
Sputtering temperature usually refers to top flooding. It is the ST below which the rewetting velocity becomes constant.
Minimum film boiling temperature corresponds to a minimum in the heat flux map obtained when the heat flux from the wall is plotted as a function of the wall temperature, mass flux, quality, etc.
Leidenfrost temperature is the ST for which the evaporation time of a quiescent droplet on a horizontal surface becomes a maximum.
Two main mechanisms have been proposed to determine the rewetting temperature. The hydrodynamic approach holds that the separation of the liquid-vapor interface from the wall lasts only as long as the vapor generation rate is sufficient to maintain a stable vapor film. According to this theory, rewetting commences due to hydrodynamic instabilities which depend on the velocities, densities, and viscosities of both phases as well as the surface tension at the liquid-vapor interface.
The thermodynamic approach assumes that the liquid can never exist beyond a maximum liquid temperature, MLT, (or superheat) which depends on the liquid properties only. Thus, a heated surface whose surface temperature exceeds this threshold cannot support liquid contact. The MLT can be determined either from the spinodal line, or from the kinetic theory of bubble nucleation in liquids.
Other approaches for defining the rewetting temperature are based on the adsorption characteristics of the system. For a static configuration, surface wettability is suitably expressed by the liquid-solid Contact Angle, θc. The knowledge of the dependency of θc on temperature enables determination of the temperature which corresponds to various contact situations, e.g., Segev and Bankoff (1980) and Olek et al. (1988a). For a dynamic configuration it is claimed that as long as the ST allows the formation of at least one monolayer of liquid molecules on the surface, wetting is possible. When the temperature is increased above a threshold value, no continuous monolayer (or close packed patches of adsorbate) can be formed, and initial spreading of the liquid over the surface will terminate.
There, is no general consensus about the effect of the various system parameters on the rewetting temperature. These effects are included in correlations for the RT, e.g., Elias and Yadigaroglu (1978).
There exists voluminous literature of rewetting models based on axial conduction, e.g., review of Elias and Yadigaroglu (1978) and Olek (1988). These models predict with partial success the rewetting velocity. Common to most of these models is the use of the upstream wet-side heat transfer coefficient and the rewetting temperature as input parameters. Some models take into account the cooling ahead of the quench front (precursory cooling), which requires an assumption of additional parameters. The weakness of these models is the arbitrariness introduced by the choice of the noted parameters [Yadigaroglu et al. (1990)]. Olek et al. (1988b) attempted to reduce the number of these rather arbitrary parameters by treating top flooding of a single rod in an unconfined geometry as a conjugate heat transfer problem. The assumptions used, however, limit the application of the model only to the particular flow configuration for which it was derived.
Most of the experimental and theoretical works on rewetting of hot surfaces have dealt with the macroscopic phenomena, such as the quench front velocity, because of its application primarily in the nuclear industry. There still does not exist a theoretical method for a precise determination of the RT with an experimental support, based on a microscopic approach.
Aksan, S. N. (1988) A Review of Large Break Loss-Of-Coolant Accident Slowdown Quench and Effect of External Thermocouples, The 3rd Int. Code Assessment and Application Program (ICAP) Specialist Meeting, Grenoble, France. March 1–4, 1988.
Barnea, Y., Elias, E., and Shai, I. (1994) Flow and heat transfer regimes during quenching of hot surfaces, Int. J. Heat Mass Trans., 37, 1441–1453. DOI: 10.1016/0017-9310(94)90146-5
Butterworth, D. and Owen, R. G. (1975) The Quenching of Hot Surfaces by Top and Bottom Flooding—A Review, Report No. AERE-R7992, AERE Harwell, Oxfordshire, England.
Carbajo, J. J. and Siegel, D. (1980) Review and comparison among the different models for rewetting in LWR’s, Nucl. Eng. Des., 58, 33–44. DOI: 10.1016/0029-5493(80)90091-6
Groeneveld, D. C. (1984) Inverted Annular and Low Quality Film Boiling: A State-of-the-Art Report, The 1st International Workshop on Fundamental Aspects of Post-Dryout Heat Transfer, Salt Lake City Utah, USA, April 1–4.
Elias, E. and Yadigaroglu, G. (1978) The reflooding phase of LOCA in PWRs, Part II, Rewetting and liquid entrainment, Nucl. Safety, 19, 160–175.
Gerweck, V. (1989) Rewetting Phenomena and their Relation to Intermolecular Forces Between a Hot Wall and the Fluid, PSI Rept. No. 42, Paul Scherrer Institute, Switzerland, December.
Olek, S., Zvirin, Y, and Elias, E. (1988a) The relation between the rewetting temperature and the liquid-solid contact angle, Int. J. Heat Mass Trans., 31, 898–902. DOI: 10.1016/0017-9310(88)90147-0
Olek, S. (1988) Analytical Models for the Rewetting of Hot Surfaces, PSI Rept. No. 17, Paul Scherrer Institute, Switzerland, October.
Segev, A. and Bankoff, G. (1980) The role of adsorption in determining the minimum film boiling temperature, Int. J. Heat Mass Trans., 23, 623–637. DOI: 10.1016/0017-9310(80)90007-1
Yadigaroglu, G., Andreoni, M., Aksan, S. N., Lewis, M. J., Analytis, G. Th., Lübbesmeyer, D., and Olek, S. (1990) Modeling of Thermohydraulic Emergency Core Cooling Phenomena, PSI Rept. No. 27, Paul Scherrer Institute, Switzerland, October.